cochrane phd writing tips

The Grumpy Economist

John Cochrane's blog

Saturday, November 22, 2014

Writing compactly, 12 comments:.

No need to write a paper if you can make your point in a tweet. Except that tweets don't build a good CV )-;

Very good advice. Unless you're being paid by the word (as with some famous 19th century authors), in which case "never use 1 word when 10 will do". :-)

In my salad days, I was a corporation lawyer. I had a client who absolutely refused to read anything I sent him that was more than one page long. Further he insisted that the page have generous margins and the writing be set forth in a series of "bullet points" (i.e. like a power point slide). And, he had zero tolerance for jargon. Keeping him happy was quite a challenge, but I learned a lot about how to be concise and communicative.

Prof Cochrane - did you, in the past, post a 4 page paper on using the language/ methodology of software programming to describe economic modelling? If so, would you mind posting a link to it on your blog? Thanks.

The advice for journalists and economic commentators working for newspapers is of course the direct opposite of the above. I.e. if you can take a simple point and turn it into a thousand words of impressive sounding hot air and waffle, you’ll be guaranteed a secure job and decent pay with some newspaper.

There are few points worth making that cannot be made in 250 words. Many concepts could have been a brilliant essay but instead became a mediocre 350 page book.

Smaller point but I used your conclusion in a discussion this fall "May economists falsely think of themselves as scientists who just “write up” research. We are not" just a typo (many) worth fixing while you're at it. Thanks for the wonderful advice!!!!

Yeah. "Fix typos before you release anything" is more good advice in the "do as I say don't do as I do" category.

Please share more content.

I am a student and that post is very useful for me.

It must be done through some efficient ideas.

Comments are welcome. Keep it short, polite, and on topic. Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.

• High School
• You don't have any recent items yet.
• You don't have any courses yet.
• You don't have any books yet.
• You don't have any Studylists yet.
• Information

Phd paper writing

Dissertation in business adm (bus x799), indiana university, students also viewed.

• D-271 note - A lecture note
• F18 K204 Syllabus Final
• Nutrition pt2 - notes
• Active Learning Template Gand D Adolescent
• Voting Theory Study Guide
• B250 W10 Reading Reflection

Related documents

• 412 Exam III Study Guide
• Wounded Knee Reading
• U.S. Response to the Holocaust
• New Deal Programs
• D271 Final - good stuff
• Leadership and Management Review

Preview text

Writing tips for ph. d. students, john h. cochrane 1 , 2, graduate school of business, university of chicago, 5807 s. woodlawn, chicago il 60637., 773 702 3059., [email protected], uchicago/fac/john.cochrane/research/papers/, june 8, 2005.

1 Always put your contact info on the front page so that people can find your paper and send

you comments! Itís the 21st century ó get a web page. If your paper is ready for a faculty member to read it, it should be on your webpage. Put the date on the paper so people know if they are reading a new version. 2 I thank Toby Moskowitz for helpful comments.

1 Organization

Figure out the one central and novel contribution of your paper. Write this down in one paragraph. As with all your writing, this must be concrete. Donít write ìI analyzed data on executive compensation and found many interesting results.î Explain what the central results are. For example, Fama and French 1992 start their abstract with: ìTwo easily measured variables, size and book-to-market equity, combine to capture the cross-sectional variation in average stock returns associated with market β, size, leverage, book-to-market equity, and earnings-price ratios.î

Distilling your one central contribution will take some thought. It will cause some pain, because you will start to realize how much youíre going to have to throw out. Once you do it, though, youíre in a much better position to focus the paper on that one contribution, and help readers to get it quickly.

Your readers are busy and impatient. No reader will ever read the whole thing from start to finish. Readers skim. You have to make it easy for them to skim. Most readers want to know your basic result. Only a few care how it is different from others. Only a few care if it holds up with different variable definitions, different instrument sets, etc.

Organize the paper in ìtriangularî or ìnewspaperî style, not in ìjokeî or ìnovelî style. Notice how newspapers start with the most important part, then fill in background later for the readers who kept going and want more details. A good joke or a mystery novel has a long windup to the final punchline. Donít write papers like that ó put the punchline right up front and then slowly explain the joke. Readers donít stick around to find the punchline in Table 12.

The vast majority of Ph. student papers and workshop presentations (not all by stu- dents!) get this exactly wrong, and we never really find out what the contribution of the paper is until the last page, the last table, and the last 5 minutes of the seminar.

A good paper is not a travelogue of your search process. We donít care how you came to figure out the right answer. We donít care about the hundreds of things you tried that did not work. Save it for your memoirs.

Most journals allow 100-150 words. Obey this limit now. The main function of the abstract is to communicate the one central and novel contribution, which you just figured out. You should not mention other literature in the abstract. Like everything else, the abstract must be concrete. Say what you find, not what you look for. Here too, donít write ìdata are analyzed, theorems are proved, discussion is made..î

Body of the paper

Your task now is to get to the central result as fast as possible. Most papers do precisely the opposite: They have a long motivation, a long literature review, a big complex model that then gets ignored, descriptive statistics, preliminary results, a side discussion or two and then finally Table 12 of ìmain estimates.î By then weíre all asleep.

Hereís the rule: There should be nothing before the main result that a reader does not need to know in order to understand the main result.

In most papers, the ìmain resultî is empirical. There may be some theory or a model, but if you (or the editor!) ask ìdoes this paper expand our knowledge of economic theory?,î the answer is ìno.î The theory is there to help understand the empirical work. Following the rule, then, the theory must be the minimum required for the reader to understand the empirical results.

Do not write a ìgeneralî model and then ìfor the empirical work, we now specialize the general shock process to an AR(1), we use only 2 firms rather than a continuum, we assume agents have quadratic utility,î etc. Work out only the specialized model that you actually take to data.

Empirical work

Start with the main result. Do not do warmup exercises, extensive data description (especially of well-known datasets), preliminary estimates, replication of othersí work. Do not motivate the specification that worked with all your failures. If any of this is really important, it can come afterwards or in an appendix.

You will mightily resist this advice. If you canít follow it, at least do not put anything before the main result that a reader does not need to know in order to understand the main result.

Follow the main result with graphs and tables that give intuition, showing how the main result is a robust feature of compelling stylized facts in the data. Follow that with limited responses to potential criticisms and robustness checks. Most of those should end up in your web appendix.

Conclusions

Really, a conclusions section should not be necessary. If you did a good job of explaining your contribution in understandable prose in the introduction, and then documenting those claims in the body of the paper, (writing in good triangular style), then saying it all over again is pointless. I tried omitting the conclusions section a few times, though, and this was too radical for editors and referees. It is true that some people skip to the conclusion to

look for the main result, but thatís because they are used to authors who donít explain it well enough in the introduction.

Thus, conclusions should be short and sweet. Do not restate all of your findings. One statement in the abstract, one in the introduction and once more in the body of the text should be enough! You can include a short paragraph or two acknowledging limitations, suggesting implications beyond those in the paper. Keep it short though ó donít write your grant application here outlining all of your plans for future research. And donít speculate; the reader wants to know your facts not your opinions.

Appendices are a great tool. Take that delicious section that has so many insightful comments on the literature, the general version of the model, the 57 robustness exercises that you did, and dump them in to an appendix. This is a good way to get them out of the paper. Eventually youíll dump them out of the appendix too.

Seriously, careful authors, referees and critics often want to document that the main result is robust to various other ways of doing things. You have to do that, but once youíve verified that it does not make that much difference and youíve found the one best way of doing things in your main result, it isnít worth space in the paper to present all the checks and variations. Appendices are a great way to solve this problem, and you can just summarize all the things you did in the paper. You can put the appendix on your and the journalís website. (ìBond risk premiaî with Monika Piazzesi is an example of a web-appendix gone wild.)

Keep it short

Keep the paper as short as possible. Every word must count. As you edit the paper ask yourself constantly, ìcan I make the same point in less space?î and ìDo I really have to say this?î Final papers should be no more than 40 pages. Drafts should be shorter. (Do as I say, not as I do!) Shorter is better.

Donít repeat things. In other words, if youíve said it once, you donít have to say it again. Most of all, it uses up extra space and readerís patience to have to see the same point made over and over again. So, once again, repetition is really a bad idea. (Get the picture?!) ìIn other wordsî is a sign of trouble. Go back and say it once, right.

General points

Follow the rule ìfirst describe what you do, then explain it, compare it to alternatives,

the variables, especially the left hand variable.

No number should appear in a table that is not discussed in the text. You donít have to mention each number separately; ìRow 1 of Table 3 shows a u-shaped patternî is ok. ìTable 5 shows summary statisticsî (period) is not ok. If itís not worth writing about in the text, itís not worth putting in the table.

Use the correct number of significant digits, not whatever the program spits out. 4. with a standard error of 0 should be 4 with a standard error of 0. Two to three significant digits are plenty for almost all economics and finance applications.

Use sensible units. Percentages are good. If you can report a number as 2 rather than 0, thatís usually easier to understand.

Good figures really make a paper come alive, and they communicate patterns in the data much better than big tables of numbers. Bad or poorly chosen figures waste a lot of space. Again, give a self-contained caption, including a verbal definition of each symbol on the graphs. Label the axes. Use sensible units. Donít use dotted line types that are invisible when reproduced. Donít use dashes for very volatile series.

Writing tips

The most important thing in writing is to keep track of what your reader knows and doesnít know. Most Ph. students assume far too much. No, we do not have the details of every paper ever written in our heads. Keep in mind what you have explained and what you have not.

The reader usually wants most of all to understand your basic point, and wonít start criticizing it before he or she understands it. Thatís behind my advice to first state and explain what you do, and save defending it and comparing it to other approaches until much later.

Use active tense. Not: ìit is assumed that τ = 3î, ìdata were constructed as follows..î Gee, I wonder who did that assuming and constructing? Search for ìisî and ìareî in the document to root out every single passive sentence.

ìIî is fine. Donít use the royal ìweî on a sole-authored paper. ìI assume that τ = 3.î ìI construct the data as follows.î If it seems like too much ìI,î you can often avoid the article altogether. For example, I think itís ok to write ìTable 5 presents estimatesî rather than ìI present estimates in Table 5î, though a purist might object to making a Table the subject of a sentence. I use ìweî to mean ìyou (the reader) and I,î and ìyouî for the reader. ìWe can see the u-shaped coefficients in Table 5î or ìYou can see the u-shaped coefficientsî is much better than ìThe u-shaped coefficients can be seenî (passive) or ìone can see the u-shaped coefficientsî (who, exactly?)

Much bad writing comes down to trying to avoid responsibility for what youíre saying. Thatís why people resort to passive sentences, ìit should be noted thatî, poor organization with literature first and your idea last, and so on. Take a deep breath, and take responsibility for what youíre writing.

Present tense is usually best. You can say ìFama and French 1993 find thatî even though 1993 was a while ago. The same goes for your own paper; describe what you find in Table 5 not what you will find in Table 5. Most importantly, though, keep the tense consistent. Donít start a paragraph in past tense and finish it in the future.

Use the normal sentence structure: subject, verb, object. Not: ìThe insurance mech- anisms that agents utilize to smooth consumption in the face of transitory earnings fluc- tuations are diverseî Instead: ìPeople use a variety of insurance mechanisms to smooth consumption..î (I also changed the starchy ìagentsî to the concrete ìpeople,î and the sim- ple ìvarietyî rather than the fancy ìdiverse.î Actually, this whole sentence probably should be dumped; it was introducing a paragraph that described the mechanisms. Itís a throat- clearing sentence that violates the rule that every sentence should mean something. The fact that people use a variety of mechanisms is not big news, the news is what the mechanisms are.)

Avoid technical jargon wherever possible.

Writing should be concrete, not abstract. (Insert concrete examples.)

Little writing tips

Donít use adjectives to describe your work: ìstriking resultsî ìvery significantî coeffi- cients, etc. If the work merits adjectives, the world will give them to you.

If you must use adjectives, donít use double adjectives. Results are certainly not ìvery novel.î

Use simple short words not big fancy words. ìUseî not ìutilize.î ìseveralî not ìdiverseî.

It is usually the case that most good writers find that everything before the ìthatî should be deleted from a sentence. Read that sentence again starting at ìEverythingî: itís true, isnít it? ìIt should be noted thatî is particularly obnoxious. Just say what you want to say. ìIt is easy to show thatî means that it isnít. Search for ìthatî in the document to get rid of these. Similarly, strike ìA comment is in order at this point.î Just make the comment. These phrases also violate the rule that each sentence should mean what it says. Is the point of the sentence really that ìit should be noted?î Or is this just a wimpy way to bring up the topic?

Clothe the naked ìthis.î ìThis shows that markets really are irrational...î This what? ìThisî should always have something following it. ìThis regression shows that....î is fine. More generally, this helps (no, that should be ìthis rule helps,î right?) you to avoid an unclear antecedent to the ìthis.î Often there are three or more things in recent memory

without saying. I donít list every single place Iíve given the workshop in the thanks. Iím not ungrateful, but the long list can get out of hand.

Donít start your paper with a cute quotation.

Donít overuse italics. (I use them far too much.) Itís best to use them only when the emphasis in a sentence would otherwise not be clear ó but maybe then you should rewrite the sentence so that the emphasis really is clear. (Who is that shouting in here?)

When describing the sign of a casual link, one direction is enough. ìWhen Jane goes up (down) on the teeter-totter, Billy goes down (up) on the other side,î the stuff in the parentheses is distracting. Add ìand vice versaî if you must.

Every sentence should have a subject, verb and object. No sentences like ìNo sentences like this.î

3 Tips for empirical work

These tips verge on ìhow to do empirical workî rather than just ìhow to write empirical work,î but in the larger picture ìdoingî and ìwritingî are not that different.

What are the three most important things for empirical work? Identification, Identifica- tion, Identification. Describe your identification strategy clearly. (Understand what it is, first!) Much empirical work boils down to a claim that ìA causes B,î usually documented by some sort of regression. Explain how the causal effect you think you see in the data is identified.

Describe what economic mechanism caused the dispersion in your right hand variables. No, God does not hand us true natural experiments very often.

Describe what economic mechanism constitutes the error term. What things other than your right hand variable cause variation in the left hand variable?

Hence, explain why you think the error term is uncorrelated with the right hand variables in economic terms. There is no way to talk about this crucial assumption unless you have done items 1 and 2!

Explain the economics of why your instruments are correlated with the right hand variable and not with the error term.

Do you understand the difference between an instrument and a control? In regressing y on x, when should z be used as an additional variable on the right hand side and when should it be an instrument for x?

Describe the source of variation in the data that drives your estimates, for every single number you present. For example, the underlying facts will be quite different as you

add fixed effects. With firm fixed effects, the regression coefficient is driven by how the variation over time within each firm. Without firm fixed effects, the coefficient is (mostly) driven by variation across firms at a moment in time.

Are you sure youíre looking at a demand curve, not a supply curve? As one way to clarify this question, ask ìwhose behavior are you modeling?î Example: Suppose you are interested in how interest rates affect housing demand, so you run the number of new loans on interest rates. But maybe when housing demand is large for other reasons, demand for mortgages (and other borrowing demand correlated with demand for mortgages) drives interest rates up. You implicitly assumed stable demand, so that an increase in price would lower quantity. But maybe the data are generated by a stable supply, so that increased demand raises the price, or some of both. Are you modeling the behavior of house purchasers or the behavior of savers (how savings responds to interest rates)?

Are you sure causality doesnít run from y to x, or from z to y and x simultaneously? Think of the obvious reverse-causality stories. Example: You can also think about the last example as causality: Do interest rates cause changes in housing demand or vice versa (or does the overall state of the economy cause both to change)?

Consider carefully what controls should and should not be in the regression. Most papers have far too many right hand variables. You do not want to include all the ìdeterminantsî of y on the right hand side.

(a) High R 2 is usually bad ó it means you ran left shoes = α + β right shoes +γprice + error. Right shoes should not be a control! (b) Donít run a regression like wage = a + b education + c industry + error. Of course, adding industry helps raise the R 2 , and industry is an important other determinant of wage (it was in the error term if you did #2). But the whole point of getting an education is to help people move to better industries, not to move from assistant burger-flipper to chief burger-flipper.

Give the stylized facts in the data that drive your result, not just estimates and p values. For a good example, look at Fama and Frenchís 1996 ìMultifactor explanations.î In the old style we would need one number: the GRS test. Fama and French show us the expected returns of each portfolio, they show us the beta of each portfolio, and they convince us that the pattern of expected returns matches the pattern of betas. This is the most successful factor model of the last 15 years .. though the GRS test is a disaster! They were successful because they showed us the stylized facts in the data.

Explain the economic significance of your results. Explain the economic magnitude of the central numbers, not just their statistical significance. Especially in large panel data sets even the tiniest of effects is ìstatistically significant.î (And when people show up with the usual 2 t statistic in large panel data sets, the effect is truly tiny!)

members will come up with stories that reverse all the signs.) Then we get some distracting preliminary results and tables and graphs of unrelated observations. More pointless discus- sion erupts; people donít know what point the speaker is trying to make and the discussion goes off in to tangents. Finally the speaker sees there is only 10 minutes to go, tells people to be quiet, and the main results go by in a big rush. Everyone is tired and confused and doesnít follow anything. I timed the finance workshop last winter quarter and not one paper got to the main results in under an hour!

Listen to the questions, all the way to the end, then count to three before answering. Yes, youíre in a rush, and yes, you think you can guess what the question will be and you know the answer. This isnít a game show, and much of the time you actually donít know what the question will be.

Keep a sheet of paper handy. You may not have a quick answer to every question, and some questions may point to good things to change in the paper.

You cannot make it too simple. Most presenters, especially Ph. D. students overestimate dramatically how much theory people can digest in one sitting, and how quickly they can memorize and digest models and results.

Speak loudly, slowly and clearly.

Thereís nothing wrong with ending early!

5 Conclusion

May economists falsely think of themselves as scientists who just ìwrite upî research. We are not; we are primarily writers. Economics and finance papers are essays. Most good economists spend at least 50% of the time they put into any project on writing. For me, itís more like 80%.

Pay attention to the writing in papers you read, and notice the style adopted by authors you admire.

I got a lot out of reading William Zinsserís On Writing Well, and D. McCloskeyís Rhetoric of Economics. I also found Glenn Ellisonís ìThe slowdown of the economics publishing processî in the JPE useful for thinking about how papers should be structured (and refereed and edited, but thatís another story).

• Multiple Choice

Course : Dissertation In Business Adm (BUS X799)

University : indiana university.